(a) You claim that only NVidia has direct 3D rendering.
Wrong, I never said "only NVidia has direct 3D rendering". I've stated nVidia provides all the options simultaneously, where other drivers do not provide all due to the various issues mentioned. I've also mentioned nVidia (besides mesa's software renderer) is the only driver that can do things like redirected rendering.
The reason so many people rave about AIGLX so much is because it allows things like Compiz - since the window contents are drawn by the X server, indirect rendering is the easiest way of doing a compositing window manager.
Easiest? I don't know. Fixing x.org instead of having to use workarounds all the time seems easier in the long term to me.
The performance impact isn't an issue since the compositing operations are really simple.
You may have not played games under AIGLX, but I have, and when the FPS drops by 30FPS because of indirect rendering, I don't consider that a non-issue.
(The issue with Compiz is that currently you can't use 3D apps at the same time reliably, since they can't render to an offscreen buffer. This is what DRI2 is intended to solve, and it's tricky to get right since it involves multiple apps with access to the same buffers. Support would be here already, but there were issues with some of the driver developers.)
I'm fully aware of these issues, which was why there are terrible AIGLX/mesa 'hacks' involved to get around this with people in the middle of it, claiming that there is no performance loss and infact you get better performance - what a load of crap.
(b) It is possible to do offscreen buffers without a proper memory manager, it's just not a good idea. (I'm unclear on why precisely). Besides, the memory manager issue was pretty much done and dusted by the time you wrote this - everyone agrees GEM is the way forward, the initial version has been merged already and will be in the next kernel/X.org releases, and drivers are in development.
Everybody agreed GEM was the way forward before, merged it, then at the last minute stripped it out and decided to rewrite it. Until it's actually part of a Linux distro, it's just vaporware to me - I have heard it all before and not only once, but numerous times. These were issues that were supposed to be resolved five years ago, I might add.
Also, I think ATI may have their own memory manager and some support for offscreen rendering/pbuffers in the closed source driver. There's just no support for using a compositing WM and OpenGL apps, since that does require DRI2.
ATi have a hack with fglrx, but they don't 'hook' into x.org, remapping the majority of functions like nVidia does.
I think it's quite reasonable for the French to want to protect this multi-trillion dollar industry. I mean, where would we be without all that great French music from companies like
How is my post troll? So far nobody has managed to give working refutable information.
So far, the most opposition I've had was stuff that was discussed, not implemented, was implemented and removed, vapourware etc. nothing that could refute the current state that X.org and various opensource projects are in.
Wait a sec, because I defend myself from being called a nVidia zealot (I'm not, I just choose the best solution on technical merit) when in reality the other person is being a opensource zealot, implying anything that isn't opensource is useless and my post is deemed flamebait?
Sure, there are different sides to the argument, but this is by no means flamebait.
You're honestly saying that Xorg doesn't support DRI? Really?
Read my journal post, DRI support is extremely limited unless you use nVidia.
You state on your 'People in the opensource community claim that there is no real performance penalty because of certain accelerated features, despite the fact 3D applications are just slowed down because they have to use non/semi-accelerated indirect rendering (depends on the driver).'
Indeed, where did I state this was Xorg developers? These are people in the opensource community, who are going around promoting Linux everywhere etc.
Because it's not done. If every company did it NVidia's way, we'd a have reimplementation of a manager manager for every driver. Are you really arguing that this is the best solution?
At the moment, nVidia is working, nothing else is, so at the moment, the best solution is nVidia.
I am talking practically, at the moment. I am fedup of waiting for things progress, most of the time they don't progress, take different turns (this problem was supposed to be solved five years ago and look, it still hasn't) etc. I live in the now.
I'm of the opinion that DRM's aren't there to be unbreakable protections anymore, but to give users pirating the game the biggest hassle possible. It may not be a problem for the tech savvy to download and implement a complicated DRM workaround, but the average user would balk at anything more complicated than overwriting executables.
If you implement a DRM system that requires genuine keys on user owned servers, genuine copies binded to a single account, allowing only one account online at a given time. Require a genuine account to get a server listing on top of it.
You've pretty much won. The pirates suddenly have to create their own darknet/server listing system, patched servers to play on. Meaning they cannot interact with genuine users.
Additionally, genuine users get the advantage of possibly permabanning specific people as each user is identified uniquely with the genuine copy of their game.
Anyone old enough to remember knows that the game company I.D. software would not have existed if people had not pirated Doom. You can still do very well with a high piracy rate if the game is any good.
I remember everyone redistributing the demo because it was practically a full game (one episode, but long enough to be a full game).
Actually, I think Doom is the only game I've ever known in my youth to not see a pirate copy of.
Some of the programs you mentioned plan on being X.org-compatible from what I understand. If so, that'd basically be the same thing as making a "new" X.org. But, it would help adoption to keep the name.
The thing is, I don't see anything wrong with the current X11 protocol. I see plenty of things wrong with the current architecture provided by x.org and xfree86.
We don't need yet another windowing system, as the current limitations we have are purely due to implementation, not by protocol design.
Of course X does direct rendering. It's called Direct Rendering Interface - DRI. And the new improved DRI2 being worked on now.
It might being worked on, but the current state is that it isn't there and it isn't available right now.
His other argument is that Xorg will never be able to have a unified memory manager... which is exactly what TTM and its successor GEM do.
You are putting words in my mouth, I never stated it will never have a unified memory manager, I was saying that there are conflicts within the management of x.org to get this done. I'm aware of Gem, are you aware of how it got implemented and stripped out a few times already?
And noone in the Xorg team claims that indirect rendering is as fast as direct rendering.
Where did I say I talked to the Xorg team?
Companies like NVidia just replace chunks of Xorg without contributing anything back.
nVidia replace chunk of Xorg to make it work properly without making their changes opensource. We can't really say "contributing anything back" because they aren't really getting anything in the first place, other than a properly working X11 setup.
Whereas its companies like Intel that actually contribute to improving X for everything - pushing a unified memory manager (TTM/GEM) into the kernel etc.
I see no evidence that Xorg would benefit from a rewrite, compared to putting that energy into working on Xorg.
I just see the fact that x.org is not in good shape at the moment when it should be. I don't care if a new server is written from scratch, huge chunks of x.org are rewritten, something needs to be done.
You don't realize that the reason it's technically superior is because it's open source...
I wouldn't be using Linux right now if I had to use only opensource software and opensource components on it. Without the proprietary bits I am using, it is technically inferior for my uses.
So when you need to trust your system, how do you check the driver for quality? For adherence to standards? How do you know the driver contains no malicious code? No incompetent code? Inspection is impossible
If I thought like you, I would never of been able to reverse engineer all the driver issues, ACPI DSDT table issues, various software issues in the past. It might be harder, but it certainly is not impossible.
And when it breaks, you have no idea why.
If I can't figure out where something breaks without the sourcecode, sourcecode is not going to be that helpful to me, is it?
I agree with the fact that binary drivers are harder to support and in most cases, people just don't bother trying when they break. But don't exaggerate the issue by claiming it's impossible.
In September, Stardock reported that Sins sold over 500,000 units: 400,000 at retail and 100,000 online. For the sake of these back-of-the-envelope calculations, I'll assume that the average retail price is $40. The online price is $40. I'll round down total sales to 500,000.
Wow, that's quite a lot.
I wouldn't pirate Stardock's games though for one reason - They have nothing I want to obtain in the first place.
I admit it, I pirated world of goo because I thought it sounded sorta cool. to be honest, it's still sitting in the torrent downloads folder and it hasnt been looked it. I guess that counts as part of the 80%, but you can hardly claim I would have bought it otherwise. I mean, shit, I havent even played with it yet and it cost me nothing. can you really claim I would have bought it if I couldnt download it for free?
The statistics they have are from people who actually played and submitted scores back online. So you wouldn't of been counted.
I keep seeing sentiments like this on the Phoronix forums and elsewere, italways seems to e Nvidia Zealotry
Wait a moment. Because I care about my hardware working better with my operating system more than the company assisting with making open source I'm a nVidia zealot?
I think you need to take a look in the mirror, right now. I'm not the one advocating or 'zealoting' (if there is such a word) that things must be opensource, I just want my stuff to work. You, on the overhand are shunning nVidia because they don't release detailed specifications on their cards and any positive mentioned about nVidia must be zealotry for this reason.
How long will it be till they release any docs for their long discontinued hardware?
I don't see why they need to.
even as crap as ATI's drivers where they did release docs for the R200 series and under cards, yet I still need the Nvidia binary blob to get the now ancient Riva TNT2 M64 in my mom's Ubuntu web browser comp, let alone my 3+ year old Gforce 6200.
So you tick a checkbox in the 'restricted-drivers' (or whatever your distribution provides for managing proprietary drivers) application, I don't see the problem?
Color me unimpressed with this latest release from Nvidia, they're the ones falling behind from their once defacto standard.
What standard? How?
the open source drivers may suck at this time, but at least they're being fed documentation , which is more then you can say for Nvidia.
Yeah... nVidia only made their driver remap huge portions of X to get proper memory mapping going so users could actually use the software and hardware properly...
Now if only Intel would release a decently powerful GPU card while holding their consistency of releasing docs and code to x.org.
Even if Intel released a decently powerful GPU card, unless x.org has a major architecture change or Intel pulls a driver hack like nVidia, it is not going to be comparable to nVidia.
When will any decent video/GPU support come to Mac OS X?
When Apple gets serious about it. Apple is the one creating the APIs on OS X for OEMs to provide, unlike in the Linux/Windows world where it goes both ways.
Here is to hoping that Wayland addresses some of these issues.
Wayland is not a new x11 implementation, it's a completely new windowing implementation, similar to Aqua as it would have widgets built into the server among other things.
Personally, I love x11, it's great - Majority of the issues currently with x.org and xfree86 are not x11 related, but architecture problems in x.org itself. A clean new implementation of x11 would probably benefit us a lot more than another Y windows, Wayward, Aqua etc.
The bulk of your argument seems to be that Nvidia's got a much more complete OpenGL implementation than does anyone else. Nevermind that almost all of it is simply code duped from their MS Windows driver, your argument is really the ages-old "if it works, then who cares if it is closed source" argument we've heard time and time again.
No, the bulk of my argument is the fact that nVidia's driver replaces the bulk of x.org. Because what x.org provides is not good enough.
Nevermind that almost all of it is simply code duped from their MS Windows driver, your argument is really the ages-old "if it works, then who cares if it is closed source" argument we've heard time and time again.
I don't use Linux because it's opensource, I use it because I find it technically superior. I have no feelings that everything should be opensource.
He had to go out and spend another ~$150 for two Gefen DVI Detectives just to enable the nvidia card to see an edid so that the driver would correctly turn on the chip's DVI transmitter.
That sucks.
Nvidia's vaunted customer support? Totally clueless and useless, they completely dropped the ball, just ignoring the issue once they realized it was more than a "did you plug in the power cord" level issue.
Sorry, I don't know of any tech company that has decent support. My own experience with all companies, including AMD and ATi is similar.
So, while it is great for you personally that Nvidia's drivers work perfectly with the hardware you own, I'm pretty sure your tune would change right quick if you had to just bend over and take it due to such a trivial bug, the kind that could easily be fixed with a single line or two of code, if you just had the source.
Not really. There are numerous issues that piss me off in every driver and os combination. It's just about finding the least amount of issues and which provides best value and yes, I take sourcecode availability into account (although if the driver is crap, generally the source is too - not that helpful). However, taking all things into account, the advantages I get from nVidia are too great to ignore.
ATI was opening up their drivers. The OSS drivers were working well, and Nvidia wasn't doing anything. Nvidia addressed their horrible Linux XRender support, and now this. I may just have to stick with Nvidia in the spring.
It is actually quite far from the truth.
You might want to read a blog post I wrote about why nVidia rocks when x.org does not. It's likely to give you more reasons to move over to nVidia over ATi.
The only thing nVidia is not doing, is making their enhancements opensource.
Lynksys, Belkin, Apple, Netgear, Draytek and D-Link (and more) which account for most of the market run Linux.
Majority of TCP routers in the world run on Cisco which use IOS. The 'routers' you have brought up aren't really considered real routers. They're home consumer devices which tend to serve as a NAT gateway to the Internet.
Slashdot Mirror
Wrong, I never said "only NVidia has direct 3D rendering". I've stated nVidia provides all the options simultaneously, where other drivers do not provide all due to the various issues mentioned. I've also mentioned nVidia (besides mesa's software renderer) is the only driver that can do things like redirected rendering.
Easiest? I don't know. Fixing x.org instead of having to use workarounds all the time seems easier in the long term to me.
You may have not played games under AIGLX, but I have, and when the FPS drops by 30FPS because of indirect rendering, I don't consider that a non-issue.
I'm fully aware of these issues, which was why there are terrible AIGLX/mesa 'hacks' involved to get around this with people in the middle of it, claiming that there is no performance loss and infact you get better performance - what a load of crap.
Everybody agreed GEM was the way forward before, merged it, then at the last minute stripped it out and decided to rewrite it. Until it's actually part of a Linux distro, it's just vaporware to me - I have heard it all before and not only once, but numerous times. These were issues that were supposed to be resolved five years ago, I might add.
ATi have a hack with fglrx, but they don't 'hook' into x.org, remapping the majority of functions like nVidia does.
nVidia provides a DRI interface that provides everything, that is what is important.
Such as?
So the unique record labels like Ed Banger Records, which currently are publishers for all the well known electronic music artists?
Like Vivendi, which own,
Island Def Jam Records
Good evening, Gentlemen!
How is my post troll? So far nobody has managed to give working refutable information.
So far, the most opposition I've had was stuff that was discussed, not implemented, was implemented and removed, vapourware etc. nothing that could refute the current state that X.org and various opensource projects are in.
Wait a sec, because I defend myself from being called a nVidia zealot (I'm not, I just choose the best solution on technical merit) when in reality the other person is being a opensource zealot, implying anything that isn't opensource is useless and my post is deemed flamebait?
Sure, there are different sides to the argument, but this is by no means flamebait.
Read my journal post, DRI support is extremely limited unless you use nVidia.
Indeed, where did I state this was Xorg developers? These are people in the opensource community, who are going around promoting Linux everywhere etc.
At the moment, nVidia is working, nothing else is, so at the moment, the best solution is nVidia.
I am talking practically, at the moment. I am fedup of waiting for things progress, most of the time they don't progress, take different turns (this problem was supposed to be solved five years ago and look, it still hasn't) etc. I live in the now.
If you implement a DRM system that requires genuine keys on user owned servers, genuine copies binded to a single account, allowing only one account online at a given time. Require a genuine account to get a server listing on top of it.
You've pretty much won. The pirates suddenly have to create their own darknet/server listing system, patched servers to play on. Meaning they cannot interact with genuine users.
Additionally, genuine users get the advantage of possibly permabanning specific people as each user is identified uniquely with the genuine copy of their game.
I remember everyone redistributing the demo because it was practically a full game (one episode, but long enough to be a full game).
Actually, I think Doom is the only game I've ever known in my youth to not see a pirate copy of.
The thing is, I don't see anything wrong with the current X11 protocol. I see plenty of things wrong with the current architecture provided by x.org and xfree86.
We don't need yet another windowing system, as the current limitations we have are purely due to implementation, not by protocol design.
It might being worked on, but the current state is that it isn't there and it isn't available right now.
You are putting words in my mouth, I never stated it will never have a unified memory manager, I was saying that there are conflicts within the management of x.org to get this done. I'm aware of Gem, are you aware of how it got implemented and stripped out a few times already?
Where did I say I talked to the Xorg team?
nVidia replace chunk of Xorg to make it work properly without making their changes opensource. We can't really say "contributing anything back" because they aren't really getting anything in the first place, other than a properly working X11 setup.
Where is it? I don't see it in Ubuntu intrepid.
I just see the fact that x.org is not in good shape at the moment when it should be. I don't care if a new server is written from scratch, huge chunks of x.org are rewritten, something needs to be done.
I wouldn't be using Linux right now if I had to use only opensource software and opensource components on it. Without the proprietary bits I am using, it is technically inferior for my uses.
Check my previous posts.
If I thought like you, I would never of been able to reverse engineer all the driver issues, ACPI DSDT table issues, various software issues in the past. It might be harder, but it certainly is not impossible.
If I can't figure out where something breaks without the sourcecode, sourcecode is not going to be that helpful to me, is it?
I agree with the fact that binary drivers are harder to support and in most cases, people just don't bother trying when they break. But don't exaggerate the issue by claiming it's impossible.
Wow, that's quite a lot.
I wouldn't pirate Stardock's games though for one reason - They have nothing I want to obtain in the first place.
The statistics they have are from people who actually played and submitted scores back online. So you wouldn't of been counted.
Maybe you should RTFA.
Wait a moment. Because I care about my hardware working better with my operating system more than the company assisting with making open source I'm a nVidia zealot?
I think you need to take a look in the mirror, right now. I'm not the one advocating or 'zealoting' (if there is such a word) that things must be opensource, I just want my stuff to work. You, on the overhand are shunning nVidia because they don't release detailed specifications on their cards and any positive mentioned about nVidia must be zealotry for this reason.
I don't see why they need to.
So you tick a checkbox in the 'restricted-drivers' (or whatever your distribution provides for managing proprietary drivers) application, I don't see the problem?
What standard? How?
Yeah... nVidia only made their driver remap huge portions of X to get proper memory mapping going so users could actually use the software and hardware properly...
Even if Intel released a decently powerful GPU card, unless x.org has a major architecture change or Intel pulls a driver hack like nVidia, it is not going to be comparable to nVidia.
When Apple gets serious about it. Apple is the one creating the APIs on OS X for OEMs to provide, unlike in the Linux/Windows world where it goes both ways.
Wayland is not a new x11 implementation, it's a completely new windowing implementation, similar to Aqua as it would have widgets built into the server among other things.
Personally, I love x11, it's great - Majority of the issues currently with x.org and xfree86 are not x11 related, but architecture problems in x.org itself. A clean new implementation of x11 would probably benefit us a lot more than another Y windows, Wayward, Aqua etc.
No, the bulk of my argument is the fact that nVidia's driver replaces the bulk of x.org. Because what x.org provides is not good enough.
I don't use Linux because it's opensource, I use it because I find it technically superior. I have no feelings that everything should be opensource.
That sucks.
Sorry, I don't know of any tech company that has decent support. My own experience with all companies, including AMD and ATi is similar.
Not really. There are numerous issues that piss me off in every driver and os combination. It's just about finding the least amount of issues and which provides best value and yes, I take sourcecode availability into account (although if the driver is crap, generally the source is too - not that helpful). However, taking all things into account, the advantages I get from nVidia are too great to ignore.
It is actually quite far from the truth.
You might want to read a blog post I wrote about why nVidia rocks when x.org does not. It's likely to give you more reasons to move over to nVidia over ATi.
The only thing nVidia is not doing, is making their enhancements opensource.
Majority of TCP routers in the world run on Cisco which use IOS. The 'routers' you have brought up aren't really considered real routers. They're home consumer devices which tend to serve as a NAT gateway to the Internet.
Where is Sarah Connor?